Combined Analyses of the VHL and Hypoxia ... - ACS Publications

Oct 27, 2015 - 786-O RCC (±VHL) cells to compare signatures that reflect hypoxia and/or loss of VHL. VHL-independent hypoxic responses, notably inclu...
2 downloads 11 Views 4MB Size
Download PDF
Article pubs.acs.org/jpr

Combined Analyses of the VHL and Hypoxia Signaling Axes in an Isogenic Pairing of Renal Clear Cell Carcinoma Cells Viktor Malec, Judy M. Coulson, Sylvie Urbé, and Michael J. Clague* Cellular and Molecular Physiology Department, University of Liverpool, Liverpool L69 3BX, United Kingdom S Supporting Information *

ABSTRACT: The loss of function of the Von Hippel−Lindau (VHL) tumor suppressor leads to the development of hypervascular tumors, exemplified by clear-cell-type renal cell carcinoma (RCC). VHL governs the adaptive responses to fluctuation of oxygen levels largely through the regulated suppression of hypoxia inducible factors (HIFs). Here, we combine proteome and phospho-proteomic analysis of isogenic 786-O RCC (±VHL) cells to compare signatures that reflect hypoxia and/or loss of VHL. VHL-independent hypoxic responses, notably include up-regulation of phosphorylation at Ser232 on the pyruvate dehydrogenase α subunit that is known to promote glycolysis. Hypoxic responses governed by VHL include up-regulation of known biomarkers of RCC (e.g., GLUT1, NDRG1) and the signaling adaptor molecule IRS-2. Notably, we also observe down-regulation of linked-components associated with the Jacobs−Stewart cycle, including the intracellular carbonic anhydrase II (CA2), which governs cellular response to CO2 fluctuations that often accompany hypoxia in tumors. Further studies indicate an unusual mechanism of control for CA2 expression that, at least in part, reflects enhanced activity of the NFκB pathway, which is associated with loss of VHL. KEYWORDS: VHL, carbonic anhydrase 2, hypoxia, clear cell renal carcinoma

■

INTRODUCTION Mutations in the VHL gene predispose individuals to renal clear cell carcinoma (RCC), the most common form of kidney cancer. The best-characterized function of VHL is to control levels of the hypoxia inducible factor (HIF) 1α and 2α subunits, which regulate complex transcriptional programs. VHL is the substrate recognition component of the VCB-Cullin 2 ubiquitin E3-ligase complex that selects prolyl-hydroxylated HIFs for degradation. Under conditions of low oxygen tension (hypoxia), HIF-α subunits accumulate owing to reduced hydroxylation and binding to VHL. They then dimerize with HIF-β and transcriptionally regulate an adaptive response.1 Genetically determined loss of VHL function is therefore proposed to echo the hypoxic response. The set of genes regulated by HIF1α or HIF2α overlap but are not identical. With respect to RCC, HIF2α is emerging as a critical driver of tumor growth. All RCC cell lines express HIF2α but not always HIF1α.2 In 786-O cells which lack VHL, HIF-2α is prominently expressed and suppressed by VHL reconstitution, whereas HIF-1α is not detectable at the protein level.3 Effects on RCC tumor formation are nonequivalent with overexpression of HIF2α promoting and HIF1α retarding tumor growth.4 However, VHL is also a multifunctional protein that possesses adaptor functions independent of the Cullin complex. These influence the cytoskeleton and extracellular matrix organization as well as HIF-independent aspects of gene transcription.5−7 © 2015 American Chemical Society

The transcriptional response to both hypoxia and VHL status has been profiled in many cell lines and clinical samples. This information provides only an indirect indication of relative protein expression levels, which are subject to differential rates of mRNA translation and protein turnover.8 Large-scale proteomic studies of RCC have so far largely been confined to analysis of clinical samples. For example, tumor samples have been compared with adjacent tissue in a semiquantitative manner. One previous study has used label-free analysis to compare two RCC cell lines, 786-O and Caki-2, which differ in HIF2α expression, and another study has compared surfaceexpressed proteins in VHL-negative UMRC2 cells with VHL transfectants.9,10 Our approach, described herein, is to directly compare isogenic cell lines; parental 786-O cells which lack VHL with those in which VHL expression has been stably reconstituted (786-O/VHL). This provides several new dimensions. First of all, we can label cells to equilibrium with amino acids containing stable isotopes of differing mass (SILAC) and directly compare proteomes by LC-MS in a quantitative manner, through the relative intensities of coeluting peptides. Second, this configuration allows us to directly compare the influence of VHL status with the effect of hypoxia on 786-O/VHL cells, the first such proteomic analysis. Third, the cellular system is amenable to parallel phosphoproReceived: July 24, 2015 Published: October 27, 2015 5263

DOI: 10.1021/acs.jproteome.5b00692 J. Proteome Res. 2015, 14, 5263−5272

Article

Journal of Proteome Research teome analysis, which we have also included in our experiments. We provide quantitative data for around 2000 proteins and a similar number of individual phosphopeptides. These reveal factors that as expected are sensitive to both VHL status and hypoxia, some of which may be novel candidate biomarkers. We also find several interesting examples that are sensitive to only one parameter, such as Collagen IV (VHL) and the phosphorylation of pyruvate dehydrogenase (hypoxia).

■

RESULTS

Proteomic Data

A schematic representation of the SILAC experiments is presented in Figure 1. Three experiments were performed in a

Figure 1. Schematic overview of experiments. (i) Three experiments were performed in a duplexed configuration with the indicated isotopic labels, simply comparing 786-O and 786-O VHL cells in normoxia. (ii) A further three experiments were performed in the indicated triplexed configuration, adding the condition of 786-O/VHL cells in hypoxia (1% O2) for 48 h. Samples were mixed immediately after lysis and then processed for phosphopeptide enrichment, using FASP, followed by strong cation exchange (SCX) chromatography and TiO2 enrichment. A parallel set of lysates (proteome) was processed by taking slices from SDS-PAGE gels and performing in-gel trypsin digestion. All six experiments allowed a direct comparison of phosphopeptides and proteins between 786-O and 786-O/VHL cells in normoxia.

Figure 2. Proteome and phosphoproteome of 786-O vs 786-O/VHL cells in normoxia. Volcano plot of (A) proteome and (B) phosphoproteomic data derived from six experiments ((i) and (ii) in Figure 1). X-axis indicates the averaged ratios of proteins or phosphopeptides (i.e., O/VHL = 786-O divided by 786-O/VHL). Red points indicate >2-fold change, blue points >1.5-fold change.

Three triplexed experiments (configuration (ii), Figure 1) allowed a direct comparison between the effect of VHL status in 786-O cells versus the influence of hypoxia upon 786-O/ VHL cells. These data are plotted in Figure 3A (proteome, 1331 proteins) and 3B (phosphoproteome, 1984 phosphopeptides from 926 proteins). 786-O/VHL cells in normoxic conditions provide a common denominator for the two ratios represented by x and y axes. We noticed that the set of proteins depleted in 786-O cells yet insensitive to 48 h of hypoxia were enriched for mitochondrial proteins. This is even more stark when we look at the distribution of 28 proteins specifically associated with oxidative phosphorylation, whereas glycolysis associated proteins show no clear partitioning (Supplementary Figure 1). Conversely, when we turn to the phosphoproteomic data, we observe that a critical regulatory phosphoserine (Ser 232) on the α-subunit of pyruvate dehydrogenase (PDHA1) is highly enriched under hypoxic conditions, but insensitive to VHL status (Figure 3B). Ser232 phosphorylation functions to inhibit the enzyme, flicking the “Warburg switch” toward

duplexed configuration which only provide a direct comparison between 786-O and 786-O/VHL cells in normoxic conditions (i). A further three experiments were performed in triplexed configuration adding the further condition of 786-O/VHL cells under hypoxic conditions (1% O2) (ii). Cell lysates from each condition were mixed at the earliest opportunity and subjected to SDS-PAGE followed by in-gel tryptic digestion and LC-MS/ MS (proteome) or processed by filter aided sample preparation (FASP), followed by enrichment of phosphopeptides before LC-MS/MS (phosphoproteome). Aggregating all six experiments allowed a quantitative comparison of 2007 proteins and 2190 phosphopeptides (from 1093 proteins) between 786-O cells ± VHL under normoxic conditions (Figure 2A,B, Supplementary Table 1,2). We have identified 26 proteins and phosphopeptides from 32 proteins which differ more than 2-fold according to VHL status (Figure 2, Supplementary Table 1 and 2), specific details of which are discussed later. 5264

DOI: 10.1021/acs.jproteome.5b00692 J. Proteome Res. 2015, 14, 5263−5272

Article

Journal of Proteome Research

normoxia, 947 of the phosphopeptides we identified came from 359 proteins which were also found in the proteome data set. Of these, 67 proteins contain at least one phosphopeptide which changes greater than 1.5-fold according to VHL status (Supplementary Table 3). The respective phosphopeptide to protein ratios, for each individual phosphopeptide (colored boxes) derived from these proteins, are depicted in Figure 4A. These indicate that many such changes in phosphopeptide ratios reflect the specific activity of phosphorylation rather than changes in protein level. Similar data for phosphopeptides sensitive to oxygen tension are shown in Figure 4B. The Golgilocalized Zinc transporter, SLC39A7 represents a straightforward example: we identify a single phosphopeptide which is highly enriched relative to nonphosphorylated peptides from the same protein (Figure 4A). Compare this with brain-specific angiogenesis inhibitor 1-associated protein 2 (BAIAP2), where only one out of five identified phosphopeptides is specifically enriched relative to protein. Protein levels of N-myc downstream regulated gene 1 (NDRG1) show large changes associated with VHL status and with hypoxia (Figure 3A). Yet, individual phosphopeptides associated with this protein, show particular patterns of enrichment or de-enrichment (Figure 3B and 4A,B), indicating hypoxic control of this protein through both expression and differential phosphorylation. Kinases are prominent as outliers in the phosphopeptide data set (Supplementary Table 2). Phosphopeptides from three kinases show a parallel depletion in hypoxia and in 786-O cells: TRAF2 and NCK interacting kinase (TNIK, 4 peptides containing phosphosites at Ser678, Ser707, Ser764, Ser769 all reduced), AP2-associated kinase 1 (AAK1, peptide containing Thr602 reduced, three others unchanged), and serum glucocorticoid regulated kinase 2 (SGK2, single peptide containing phosphosite at Thr68). Phosphofructokinase, platelet (PFKP) phosphopeptide is enriched in hypoxia and in 786-O/VHL cells; however, the specific activity of phosphorylation is not increased significantly, as indicated by similar enrichment in the proteomic data set. One prominent kinase substrate showing marked enhancement of multiple phosphopeptides in hypoxia and corresponding sensitivity to VHL status is the signaling adaptor subunit IRS-2 (Figure 3B). As this was absent from our proteome data, we have monitored its expression levels by Western blotting (Figure 3C). In this case, the phosphopeptide enrichment strategy has indirectly led us to identify a protein whose levels increase either upon hypoxia or loss of VHL, rather than reflecting a change in the specific activity of phosphorylation.

Figure 3. Correlation between VHL status and response to hypoxia. Plot of data derived from three triplexed experiments ((ii) in Figure 1). X-axis indicates the averaged ratios of (A) proteins or (B) phosphopeptides (i.e., O/VHL = 786-O divided by 786-O/VHL). Yaxis indicates corresponding ratios 786-O/VHL cells in hypoxia versus normoxia. Red points indicate >2-fold change, blue points >1.5-fold change. Squares points indicate common changes between hypoxia and lack of VHL (top right and bottom left quadrants), triangles indicate opposite changes, circles are points which respond to a single variable. (C) Western blot for IRS-2 indicating changes in protein expression levels, which reflect enrichment of phosphopeptides seen in (B). N = normoxia.

Control of CA2 Expression

Among proteins that are down-regulated in the face of VHL loss or hypoxia, we decided to further examine carbonic anhydrase II (CA2). CA2 is a very active enzyme (turnover number ∼106 sec−1 mol−1) that is ubiquitously expressed but particularly abundant in the kidney, where it accounts for 95% of CA activity.13 Two carbonic anhydrases CA9 and CA12 are up-regulated in a variety of cancers.14−16 CA9 is strongly expressed under hypoxic conditions due to transcriptional activation by HIF1.17 However, both of these are plasma membrane associated trans-membrane proteins, with an extracellular active site. The cytosolic enzyme CA2 has received scant attention with respect to tumor biology and control of expression.

glycolysis by reducing conversion of pyruvate to acetyl coA.11,12 Our data show that phosphorylation at an alternative regulatory site (S293) in PDHA1 is unaffected by hypoxia (Supplementary Table 2). Measured phosphopeptide ratios can vary according to the specific activity of phosphorylation or as a consequence of changing levels of protein, from which the phosphopeptides are derived. For example, in comparing 786-O cells ± VHL under 5265

DOI: 10.1021/acs.jproteome.5b00692 J. Proteome Res. 2015, 14, 5263−5272

Article

Journal of Proteome Research

Figure 4. Specific activity of individual phosphopeptides. Collation of VHL or hypoxia-sensitive phosphopeptides which have corresponding estimates for total protein based on nonphosphorylated peptides. Each colored box indicates averaged ratios of individual phosphopeptide over protein levels in (A). 786-O cells normalized to 786-O/VHL cells under normoxic conditions. A log2 value of 2 is equivalent to a 4-fold increase in specific phosphorylation activity in 786-O cells relative to 786-O/VHL cells. (B) Comparative ratios for 786-O/VHL cells in hypoxia normalized to normoxia. The criterion for inclusion in the lists is that at least one phosphopeptide per protein changes more than 1.5-fold.

constitutively expressing Renilla luciferase gene, which were cotransfected into 786-O cells. 786-O cells show around 4-fold enhanced NFκB activity under basal conditions compared to 786-O/VHL (Figure 6A), whereas no differential signal was seen with a control plasmid lacking NFκB binding sites (data not shown). Specificity of the construct for NFκB signaling was further confirmed by knock-down of the NFκB negative regulator, IκBα, leading to an increase of luciferase activity in both cell lines (Figure 6B). IκBα plays a central role in regulation of NFκB (p65-p50) activity which may be tuned further by various physiological stimuli such as hypoxia. Accordingly, levels of the signature 36kD IκBα band as visualized by Western blotting are reduced in 786-O compared with 786-O/VHL cells, and treatment with hypoxia leads to a reduction in both cell lines (Figure 6C, left panel). Note also the enhancement of a higher molecular weight form under hypoxic conditions, the significance of which we have not directly pursued in this study. Both molecular weight forms of IκBα are sensitive to treatment by TNFα and can be rescued from TNFα-induced loss by IKK-1,2 inhibitor VII but not by IKK-2 inhibitor sc-514 (Figure 6C, right panel). We next assessed if such direct manipulation of the NFκB pathway using

We confirmed these VHL- and hypoxia-sensitive changes in CA2 by Western blotting and could ascribe them to VHLdependent regulation of CA2 gene expression, as corresponding changes in mRNA levels were determined by qRT-PCR (Figure 5A−D). Confirmation of this hypothesis is provided by the reversal of CA2 protein and mRNA elevation by siRNAmediated suppression of VHL in 786-O/VHL cells (Figure 5E,F). However, attempts to rescue CA2 protein and mRNA levels in normoxic 786-O cells and hypoxic 786-O/VHL cells by siRNA depletion of HIF-2α were unsuccessful. Only one out of four individual siRNA oligonucleotides led to an increase in CA2, despite comparable levels of depletion in each case (data not shown). Key Role of NFκB Pathway in the Regulation of CA2 Expression

We wondered then if an alternative signaling pathway controlled by VHL may be responsible for control of CA2 levels. Prompted by previous reports,18,19 we compared the level of canonical NFκB signaling between 786-O and 786-O/ VHL cells using a dual-reporter gene assay. This utilized constructs containing either a NFκB (p65-p60) sensitive enhancer element for firefly luciferase expression or a 5266

DOI: 10.1021/acs.jproteome.5b00692 J. Proteome Res. 2015, 14, 5263−5272

Article

Journal of Proteome Research

Figure 5. Expression of CA2 in 786-O cells. (A) Western blot analysis of CA2 and VHL in different cell lines: Hela, U2OS and 786-O parental cells (VHL-negative) or expressing VHL (786-O/VHL). (B) Real-time RT-PCR analysis of CA2 in 786 cells. (C) CA2 expression under normoxic (N) or hypoxic (H48, 48 h of 1% CO2) conditions. (D) Real-time RT-PCR analysis of CA2 in 786-O/VHL cells incubated in normoxia (N) and hypoxia (H32, 32 h of hypoxia). For (B) and (D), the CA2 mRNA level in 786-O/VHL is expressed relative to that in 786-O cells. (E) Western blot analysis of CA2 in 786 cells transfected with control siRNA (siC), VHL siRNA (siVHL) or not transfected (−). (F) Real-time RT-PCR of CA2 in 786 cells transfected with siC, siVHL or nontransfected. For (B), (D), and (F), the CA2 mRNA level in 786-O/VHL is expressed relative to that in 786-O cells. Error bars represent STD of the mean for three biological replicates. All siRNAs were transfected at 40 nM final concentration for 2 days.

ylation that can be directly compared with the proteome data. There is an inherent bias in large-scale proteomic experiments toward high-abundance proteins. This has advantages for the discovery of useful biomarkers, as on average, they will be correspondingly easier to detect with antibodies. The phosphoproteomic data exhibits a very strong overlap in signatures between loss of VHL and hypoxia, with a small number of divergent outliers. This is also manifest at the proteomic level for those proteins up-regulated in cells lacking VHL. However, a large cohort of down-regulated proteins is also identifiable which do not respond to hypoxia (at least not within the 48 h of the experiment). The iron−sulfur domaincontaining outer-mitochondrial membrane protein CISD1, provides a unique case within the proteome data. It responds in an opposite manner to these two variables; enriched under hypoxia yet de-enriched upon loss of VHL.

TNFα would also influence CA2 expression. Acute activation by TNFα stimulation leads to depletion of CA2 and reduced mRNA levels in both cell lines (Figure 7A,B). The IKK-1,2 inhibitor VII abrogates this effect of TNFα (Figure 7A). Our data suggest that CA2 levels can at least in part be determined by the activity of the NFκB pathway, which itself is sensitive to hypoxia and under the control of the tumor suppressor gene VHL.

■

DISCUSSION

Overview of Proteomic Data

We have provided the most comprehensive proteomic data set available that specifically reports the influence of VHL status in a model cell line. Moreover, we have been able to examine the correlation with hypoxia, which is not possible with tumorderived samples. Finally, we have provided the first phosphoproteomic analysis of VHL- and hypoxia-driven phosphor5267

DOI: 10.1021/acs.jproteome.5b00692 J. Proteome Res. 2015, 14, 5263−5272

Article

Journal of Proteome Research

experiment, suggesting that this may be a longer-term adaptive change. In distinction, the loss of two components of the iron storage protein ferritin (FTH1, FTL) is seen both with loss of VHL or hypoxia. We also observe the coordinated loss or change in phosphorylation status of three factors critical to the control of acid−base balance in the kidney proximal tubule. In addition to proteomic changes already detailed for CA2, parallel reductions in phosphopeptides derived from sodium/bicarbonate cotransporter 4 (SLC4A4) are also seen. However, the Na(+)/H(+) exchange regulatory factor 3 (PDZK1) is distinct in being sensitive at the proteome level to VHL status but is insensitive to hypoxia in VHL-expressing cells. Note that CA2 and SLC4A4 bind directly to each other, forming a metabolon or physically associated complex of proteins in a sequential metabolic pathway.25 Loss of function of either can lead to renal acidosis.23,24 Moreover, these observations provide an intriguing addendum to the well-characterized overexpression of the extracellular carbonic anhydrases CA9 and CA12 in other RCC settings. In principle, both types of change in CA expression could exert a similar influence on the Jacobs− Stewart cycle, the process by which bicarbonate is cycled in and out of the cell with net proton extrusion.26 However, a recent study has suggested that intracellular CA2 is not obligatory for supporting acid extrusion, rather that low CA2 activity serves to buffer pHi from fluctuations in extracellular pCO2.27 Loss of VHL has also been associated with defects in extracellular matrix (ECM) assembly, at least some of which are proposed to be HIF-independent.28−30 We find reduced levels of fibronectin (FN1) and collagen IV α1 and α2 subunits consistent with previous studies of RCC cells. Moreover, the collagen subunits as well as two other proteins associated with collagen biosynthesis (the collagen-specific molecular chaperone SERPINH1 and the prolyl hydroxylase LEPREL1) are deenriched following VHL loss, yet are insensitive to hypoxic conditions. Thus, our data suggest a wider coordination of ECM assembly factors than hitherto appreciated and support a previous suggestion that links VHL to hypoxia-independent collagen IV network assembly.31

Figure 6. NFκB activity in 786-O cells. The relative activity of NFκB was assessed by the ratio of luminescence generated by firefly luciferase (NFκB sensitive promoter) and Renilla luciferase (constitutively active promoter). The relative activity of NFκB was normalized to that in 786-O/O or control siRNA. Error bars represent STD of the mean for three biological replicates. (A) NFκB activity in 786-O and 786-O/ VHL cells, (B) NFκB activity in 786-O and 786-O/VHL transfected with siC or IκBα siRNA (siIκBα). (C) Left panel: Western blot analysis of IκBα in 786-O or 786-O/VHL cells subjected to hypoxia for 48 h (H48) or kept in normoxia (N). HIF-2α and CA2 blots are shown as positive/negative controls for hypoxic conditions. Right panel: Western blot analysis of IκBα in 786-O/VHL cells incubated for 48 h under hypoxia (H48) and treated for 48 h with vehicle (dimethyl sulfoxide, DMSO), DMSO with TNF-α (10 ng/mL) (DMSO+TNFα), IKK-1,2 inhibitor VII (0.2 μM) with TNF-α (VII+TNF-α), IKK-2 inhibitor sc-514 (12 μM) with TNF-α (sc-514+TNF-α) or nontreated (−).

Up-Regulated in VHL-Negative Cells

We have looked for overlap of our data set with previous protein and transcriptomic profiling studies. Our most enriched proteins in VHL-negative cells have been previously associated with renal carcinoma. Aberrant overexpression of NDRG1 in renal tumors has been previously reported and accumulation of a nuclear form is associated with good prognosis.32 The glucose transporter SLC2A1 (GLUT1), an established HIF-1 target gene, is similarly up-regulated in renal tumors as judged by mRNA quantitation, providing a basis for the synthetic lethality of glucose uptake transporters with loss of VHL.33−37 In distinction to the preceding examples, some proteins were not detected in our triplexed experiments (that include a hypoxic condition) but were nevertheless found to be highly enriched in VHL-negative cells in our duplexed experiments (786-O vs 786-O/VHL). The most prominent, LCP1/Plastin, is up-regulated in kidney tumors compared with adjacent normal tissue.38 The next two most highly enriched proteins in VHL-negative cells (786-O), C15orf48/NMES1 and lysine phosphatidycholine acyltransferase 1 (LPCAT1) have not hitherto been associated with renal cancer. 786-O cells are null for PTEN, the loss of which has been shown to be associated with enhanced LPCAT1 expression and poor

Down-Regulated in Hypoxia or VHL-Negative Cells

Transcriptional analyses have identified many genes to be down-regulated under hypoxic conditions in a HIF-dependent manner. However, suppression is proposed to be exerted indirectly through HIF-dependent expression of transcriptional repressors and miRNAs.20,21 Our list of proteins that are reduced by loss of VHL is highly enriched for mitochondrial and respiratory chain proteins. This is consistent with results using RCC4 cells ± VHL, where loss of VHL led to reduced mitochondrial mass and oxygen consumption in a HIFdependent manner.22 However, in our hands, mitochondrial mass appears insensitive to hypoxia over the time course of the 5268

DOI: 10.1021/acs.jproteome.5b00692 J. Proteome Res. 2015, 14, 5263−5272

Article

Journal of Proteome Research

Figure 7. Direct manipulation of the NFκB pathway influences CA2 expression. (A,B) Acute activation by TNFα stimulation leads to depletion of CA2 and reduced mRNA levels in both cell lines, which is abrogated by the IKK-1,2 inhibitor VII.

prognosis in prostate cancer.39 Our data suggest that loss of VHL amplifies this effect and that LPCAT1 is therefore a potential marker for such double jeopardy. We identify multiple phosphopeptides from the signaling adaptor molecule IRS-2 responsive to VHL status and hypoxia but then find that this ratiometric change actually reflects increased protein levels. Consistent with our findings, IRS-2 has previously been shown to be up-regulated by a HIF-1 dependent transcriptional mechanism in breast cancer cell lines.40 IRS-2 is known to promote glycolysis by increasing the surface expression of GLUT1.41 Thus, in kidney, just as in breast cells, VHL control of HIF-2α leads to the coordinated up-regulation of glucose transporter (see above) together with a master-controller of its location. A further indicator of a coordinated shift to glycolysis is discussed below.

regulated by hypoxia in a HIF-1α-dependent manner in mouse embryo fibroblasts and renal clear cell RCC4 cells.42,43 Comparison with Transcriptome

There are striking parallels between the present proteomic data set and a previous transcriptional profile of a derivative metastatic 786-O clone, which shows 100-fold enrichment in lung colonizing activity (see supplementary Figure 2). Following reconstitution of VHL in this clone, mRNA for LCP1 and IRS-2 was found to be reduced 14- and 3.6-fold respectively, NDRG1 and GLUT1 (SLC2A1) reduced ∼2-fold and many other mRNA/protein pairs show a similar correspondence as annotated in supplementary Figure 2.44 They also find that CA2 mRNA levels are increased following VHL reconstitution, reflecting our own findings. There are also several instances where we find cellular levels of proteins changing according to VHL status but there is no corresponding change in mRNA levels according to the reported array data. These include many mitochondrial proteins, ferritin (FTH1) and SERPINH1 which are all increased and Synaptophysin Like 1 (SYPL1) and Proteasomal Maturation Protein (POMP) which decrease, following VHL reintroduction. Note that SYPL1 and POMP are insensitive to oxygen tension, suggesting they may be direct substrates of VHL.

Hypoxia-Sensitive, VHL-Insensitive Factors

Although there is a strong correlation between sensitivity to hypoxia and loss of VHL in both proteome and phosphoproteome data sets, we have identified several examples where this relationship is uncoupled. PDZK1 Interacting Protein (PDZK1IP1/MAP17) is increased in hypoxia but is not sensitive to VHL status. Note that this is the reciprocal behavior to its interacting protein PDZK1 (see above), which is VHL-sensitive but hypoxia-insensitive. Phosphorylation at Ser232 on the pyruvate dehydrogenase α subunit (PDHA1) is highly up-regulated under hypoxic conditions but insensitive to VHL status. This is one of three phosphorylation sites that are substrates of pyruvate dehydrogenase kinase (PDK) enzymes 1−4. All sites are inhibitory and hence provide a means to further shift cells toward glycolysis, manifesting the Warburg effect, by excluding pyruvate from mitochondrial consumption. Ser232 is an exclusive site for PDK1 while the other sites can be phosphorylated by multiple PDK enzymes.11 One of these (Ser293) is also found in the phosphoproteomic analysis but is insensitive to either hypoxia or VHL status. PDK1 is a target gene for HIF-1 and is up-

VHL/NFκB and Repression of CA2

Upon establishing that VHL-dependent expression of CA2 is controlled at the transcriptional level, it may first seem likely that one of the hypoxia-inducible factors is an intermediary molecule, acting for example, by up-regulation of a repressor molecule. In 786-O cells the relevant isoform is HIF-2α, which is itself repressed following reconstitution of VHL expression.2,4,45 However, siRNA experiments and pharmacological inhibition of HIF-2α translation46 did not support this interpretation (not shown). It has been previously reported that NFκB signaling is up-regulated in cells with compromised VHL expression19,47,48 and during hypoxia.18,49,50 Although the 5269

DOI: 10.1021/acs.jproteome.5b00692 J. Proteome Res. 2015, 14, 5263−5272

Article

Journal of Proteome Research

the Bio-Rad iQ5 cycler with iQ SYBR Green supermix (BioRad Laboratories Ltd., UK). Cycling conditions were: 95 °C for 15 min (denaturation), followed by 40 cycles of 95 °C for 10 s (denaturation), 58 °C for 30 s (annealing), 72 °C for 30 s (extension). For mRNA quantification, the target gene was normalized to β-actin mRNA and expression in the treated group shown relative to that in the control group (2−ΔCt).

mechanistic details of this VHL-NFκB axis have not been fully elucidated, the up-regulation of NFκB in RCC VHL− cells has been linked to the malignant phenotype, particularly by increasing resistance to apoptosis.47,51,52 In line with this, our own reporter assays confirmed increased NFκB signaling in VHL− compared with VHL+ cells. Furthermore, we have been able to show a strong influence of NFκB signaling upon CA2 expression by stimulation with TNF-α and pharmacological inhibition of IKKs. Exposure of 786-O cells to hypoxia also mirrors the effects of TNF-α stimulation, by reducing levels of IκBα and CA2. The effect of TNF-α on CA2 protein or mRNA levels is not rapid in either case. This suggests that control is not exerted immediately at the level of the CA2 gene transcription but represents an adaptive change to enhanced NFκB signaling. In summary, this work presents the first direct comparison between the effect of hypoxia and a pre-eminent governor of oxygen sensing, the tumor suppressor VHL. We confirm several findings made in scattered reports related to RCC, which are now presented as components of a global and precisely defined signature. Furthermore, we highlight additional candidate biomarkers. Our finding that the intracellular carbonic anhydrase, CA2, is repressed under hypoxic/VHL- conditions, highlights a novel output of NFκB signaling, which is likely to impact upon the responsiveness of cancer cell to fluctuations in extracellular pCO2.27 Such fluctuations are associated with unstable perfusion in tumors, which also represents the underlying cause of hypoxia.

■

NFkB Reporter Gene Assay

Cells were cotransfected with a reporter gene construct containing five NFκB (p65-p50)-sensitive enhancer repeats in the promoter of the Firefly luciferase gene (cat. no. 219078, Stratagene, Agilent Technologies, Santa Clara, CA) and a construct expressing Renilla luciferase gene under the control of the constitutively active thymidine kinase (TK) promoter (cat. no. E2241, Promega, Madison,WI). As a negative control pCIS-CK plasmid (cat. no. 219090, Stratagene, Agilent Technologies, Santa Clara, CA) containing the firefly luciferase reporter gene but no NFkB binding sites was used. Firefly and Renilla luciferase constructs were transfected in a ratio of 50 to 1. Transfection was performed in full cell culture medium using GeneJuice (cat. no. 70967, Merck Millipore, Feltham, UK). NFκB activity expressed in relative luminescence units (RLU) was measured by dual-luciferase assay (cat. no. E1980, Promega, Madison, WI) using GloMax-Multi Microplate Multimode Reader (Promega, Madison, WI). Normalized RLU represents the ratio between luminiscence of Firefly luciferase and Renilla luciferase. Normalized RLU of 786-O cells or control siRNA were set to 1.

MATERIALS AND METHODS

Sample Preparation for Mass Spectrometry

Cell Culture

Phosphopeptide enrichment was achieved by filter-aided sample preparation (FASP) of mixed cell lysates54 followed by fractionation using strong cation exchange (SCX) chromatography and TiO2-based phosphopeptide isolation as described previously.55,56 Samples for quantitative proteome analyses were prepared in parallel by resolving a 50 μg aliquot of mixed cell lysates on a 4−12% NuPAGE gel (Invitrogen). Gel lanes were then cut into 48 bands, in-gel digested overnight at 37 °C with trypsin (4 ng/μL working concentration; Trypsin GOLD, sequencing grade, Promega), dried and redissolved in 0.05% TFA prior to LC-MS/MS analysis.

Isogenic clear-cell renal cell carcinoma cells expressing (786-O/ VHL) or lacking (786-O) the human VHL gene were a kind gift from Dr. Bottaro, NCI, Bethesda, MD.53 Hygromycin B (500 μg/mL) was used for routine selection of 786-O/VHL cells. All cell lines were cultured in D-MEM supplemented with 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin at 37 °C, 5% CO2. siRNA Transfection

Cells were transfected with target-specific or nontargeting siRNA oligonucleotides at 40 nM final concentration using RNAiMAX transfection reagents (Life Technologies Ltd., U.K.) in serum free medium. Six hours after transfection the serum free medium was replaced with full medium. The siRNA used in experiments were as follows: siGenome SMART pool HUMAN CA2, HIF2α, VHL, IκBα, nontargeting siRNA #2, nontargeting siRNA #1 (all from Dharmacon, Thermo Fisher Scientific, Waltham, MA), AllStars Negative Control -1027281 (Qiagen GmbH, Hilden, Germany).

LC-MS/MS and Data Processing

Five microliters of each sample were fractionated by nanoscale C18 high performance liquid chromatography (HPLC) on a Waters nanoACQUITY UPLC system coupled to an LTQOrbitrap XL mass spectrometer. Peptides were resolved using a 25 cm × 75 μm column (BEH-C18 Symmetry; Waters Corporation) using a 21 min linear gradient of 0 to 37.5% acetonitrile in 0.1% formic acid, flow rate 400 nL min−1, column temperature, 65 °C. MS survey scans were acquired in the Orbitrap (R = 30 000; m/z range 300−2000) and MSMS on the top 5 multiply charged ions in the linear quadrupole ion trap (LTQ) after fragmentation using collision-induced dissociation (30 ms at 35% energy). Raw MS peak list files from each experimental configuration were searched against the human IPI database (version 3.77) using the Andromeda search engine and processed and statistically evaluated with the MaxQuant (version 1.2.2.5) and plotted with JMP software.57 Peptide and protein false discovery rates were set to 0.01.

Real-Time PCR

Cells were washed with PBS and RNA was isolated using RNeasy Mini Kit (Qiagen GmbH, Hilden, Germany) according to the company’s instructions. Reverse transcription was performed with 1 μg of RNA using QuantiTect Reverse Transcription Kit (Qiagen GmbH, Hilden, Germany). The cDNA templates in the real-time PCR reactions used the following primers: CA2, forward 5′-GAC AAA GCA GTG CTC AAG GGA-3′, reverse 5′-CCA AGT GAA GTT CTG CAG CA-3′; β-actin, forward 5′-CAC CTT CTA CAA TGA GCT GCG TGTG-3, reverse 5′-ATA GCA CAG CCT GGA TAG CAA CGT AC-3′. Real-time PCR was performed using 5270

DOI: 10.1021/acs.jproteome.5b00692 J. Proteome Res. 2015, 14, 5263−5272

Article

Journal of Proteome Research

■

(11) Holness, M. J.; Sugden, M. C. Regulation of pyruvate dehydrogenase complex activity by reversible phosphorylation. Biochem. Soc. Trans. 2003, 31, 1143−1151. (12) Schulze, A.; Downward, J. Flicking the Warburg switch-tyrosine phosphorylation of pyruvate dehydrogenase kinase regulates mitochondrial activity in cancer cells. Mol. Cell 2011, 44, 846−848. (13) Purkerson, J. M.; Schwartz, G. J. The role of carbonic anhydrases in renal physiology. Kidney Int. 2007, 71, 103−115. (14) Pastorek, J.; Pastorekova, S.; Callebaut, I.; Mornon, J. P.; Zelnik, V.; Opavsky, R.; Zat’ovicova, M.; Liao, S.; Portetelle, D.; Stanbridge, E. J.; et al. Cloning and characterization of MN, a human tumorassociated protein with a domain homologous to carbonic anhydrase and a putative helix-loop-helix DNA binding segment. Oncogene 1994, 9, 2877−2888. (15) Tureci, O.; Sahin, U.; Vollmar, E.; Siemer, S.; Gottert, E.; Seitz, G.; Parkkila, A. K.; Shah, G. N.; Grubb, J. H.; Pfreundschuh, M.; Sly, W. S. Human carbonic anhydrase XII: cDNA cloning, expression, and chromosomal localization of a carbonic anhydrase gene that is overexpressed in some renal cell cancers. Proc. Natl. Acad. Sci. U. S. A. 1998, 95, 7608−7613. (16) Neri, D.; Supuran, C. T. Interfering with pH regulation in tumours as a therapeutic strategy. Nat. Rev. Drug Discovery 2011, 10, 767−777. (17) Ivanov, S. V.; Kuzmin, I.; Wei, M. H.; Pack, S.; Geil, L.; Johnson, B. E.; Stanbridge, E. J.; Lerman, M. I. Down-regulation of transmembrane carbonic anhydrases in renal cell carcinoma cell lines by wild-type von Hippel-Lindau transgenes. Proc. Natl. Acad. Sci. U. S. A. 1998, 95, 12596−12601. (18) Culver, C.; Sundqvist, A.; Mudie, S.; Melvin, A.; Xirodimas, D.; Rocha, S. Mechanism of hypoxia-induced NF-kappaB. Mol. Cell. Biol. 2010, 30, 4901−4921. (19) Pantuck, A. J.; An, J.; Liu, H.; Rettig, M. B. NF-kappaBdependent plasticity of the epithelial to mesenchymal transition induced by Von Hippel-Lindau inactivation in renal cell carcinomas. Cancer Res. 2010, 70, 752−761. (20) Mole, D. R.; Blancher, C.; Copley, R. R.; Pollard, P. J.; Gleadle, J. M.; Ragoussis, J.; Ratcliffe, P. J. Genome-wide association of hypoxia-inducible factor (HIF)-1alpha and HIF-2alpha DNA binding with expression profiling of hypoxia-inducible transcripts. J. Biol. Chem. 2009, 284, 16767−16775. (21) Kulshreshtha, R.; Ferracin, M.; Wojcik, S. E.; Garzon, R.; Alder, H.; Agosto-Perez, F. J.; Davuluri, R.; Liu, C. G.; Croce, C. M.; Negrini, M.; Calin, G. A.; Ivan, M. A microRNA signature of hypoxia. Molecular and cellular biology 2007, 27, 1859−1867. (22) Zhang, H.; Gao, P.; Fukuda, R.; Kumar, G.; Krishnamachary, B.; Zeller, K. I.; Dang, C. V.; Semenza, G. L. HIF-1 inhibits mitochondrial biogenesis and cellular respiration in VHL-deficient renal cell carcinoma by repression of C-MYC activity. Cancer Cell 2007, 11, 407−420. (23) Gross, E.; Pushkin, A.; Abuladze, N.; Fedotoff, O.; Kurtz, I. Regulation of the sodium bicarbonate cotransporter kNBC1 function: role of Asp(986), Asp(988) and kNBC1-carbonic anhydrase II binding. J. Physiol. 2002, 544, 679−685. (24) Golembiewska, E.; Ciechanowski, K. Renal tubular acidosis– underrated problem? Acta Biochimica Polonica 2012, 59, 213−217. (25) Sterling, D.; Reithmeier, R. A.; Casey, J. R. A transport metabolon. Functional interaction of carbonic anhydrase II and chloride/bicarbonate exchangers. J. Biol. Chem. 2001, 276, 47886− 47894. (26) Potter, C. P.; Harris, A. L. Diagnostic, prognostic and therapeutic implications of carbonic anhydrases in cancer. Br. J. Cancer 2003, 89, 2−7. (27) Hulikova, A.; Aveyard, N.; Harris, A. L.; Vaughan-Jones, R. D.; Swietach, P. Intracellular carbonic anhydrase activity sensitizes cancer cell pH signaling to dynamic changes in CO2 partial pressure. J. Biol. Chem. 2014, 289, 25418−25430. (28) Ohh, M.; Yauch, R. L.; Lonergan, K. M.; Whaley, J. M.; Stemmer-Rachamimov, A. O.; Louis, D. N.; Gavin, B. J.; Kley, N.; Kaelin, W. G., Jr.; Iliopoulos, O. The von Hippel-Lindau tumor

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jproteome.5b00692. Supplementary data and figure/table captions; distribution of mitochondrial proteins in the proteomic data set; comparison of proteomic data with published transcriptomic data (PDF) Collated protein ratios (XLSX) Collated phosphopeptide ratios (XLSX) Individual phosphopeptide vs protein ratios (XLSX)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Phone: 44 151 794 5308. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by a Cancer Research U.K. project grant (C17689/A11806) to M.C. We thank Dr. Bottaro (NIH) for his generous gift of cell lines. We thank Ian Mills and Violaine See for helpful advice and discussion.

■

REFERENCES

(1) Baldewijns, M. M.; van Vlodrop, I. J.; Vermeulen, P. B.; Soetekouw, P. M.; van Engeland, M.; de Bruine, A. P. VHL and HIF signalling in renal cell carcinogenesis. Journal of pathology 2010, 221, 125−138. (2) Shinojima, T.; Oya, M.; Takayanagi, A.; Mizuno, R.; Shimizu, N.; Murai, M. Renal cancer cells lacking hypoxia inducible factor (HIF)1alpha expression maintain vascular endothelial growth factor expression through HIF-2alpha. Carcinogenesis 2007, 28, 529−536. (3) Lau, K. W.; Tian, Y. M.; Raval, R. R.; Ratcliffe, P. J.; Pugh, C. W. Target gene selectivity of hypoxia-inducible factor-alpha in renal cancer cells is conveyed by post-DNA-binding mechanisms. Br. J. Cancer 2007, 96, 1284−1292. (4) Raval, R. R.; Lau, K. W.; Tran, M. G.; Sowter, H. M.; Mandriota, S. J.; Li, J. L.; Pugh, C. W.; Maxwell, P. H.; Harris, A. L.; Ratcliffe, P. J. Contrasting properties of hypoxia-inducible factor 1 (HIF-1) and HIF2 in von Hippel-Lindau-associated renal cell carcinoma. Mol. Cell. Biol. 2005, 25, 5675−5686. (5) Li, M.; Kim, W. Y. Two sides to every story: the HIF-dependent and HIF-independent functions of pVHL. J. Cell Mol. Med. 2011, 15, 187−195. (6) Frew, I. J.; Krek, W. pVHL: a multipurpose adaptor protein. Sci. Signaling 2008, 1, pe30. (7) Linehan, W. M.; Rubin, J. S.; Bottaro, D. P. VHL loss of function and its impact on oncogenic signaling networks in clear cell renal cell carcinoma. Int. J. Biochem. Cell Biol. 2009, 41, 753−756. (8) Schwanhausser, B.; Busse, D.; Li, N.; Dittmar, G.; Schuchhardt, J.; Wolf, J.; Chen, W.; Selbach, M. Global quantification of mammalian gene expression control. Nature 2011, 473, 337−342. (9) Aggelis, V.; Craven, R. A.; Peng, J.; Harnden, P.; Schaffer, L.; Hernandez, G. E.; Head, S. R.; Maher, E. R.; Tonge, R.; Selby, P. J.; Banks, R. E. VHL-dependent regulation of a beta-dystroglycan glycoform and glycogene expression in renal cancer. Int. J. Oncol. 2013, 43, 1368−1376. (10) Nagaprashantha, L. D.; Talamantes, T.; Singhal, J.; Guo, J.; Vatsyayan, R.; Rauniyar, N.; Awasthi, S.; Singhal, S. S.; Prokai, L. Proteomic analysis of signaling network regulation in renal cell carcinomas with differential hypoxia-inducible factor-2alpha expression. PLoS One 2013, 8, e71654. 5271

DOI: 10.1021/acs.jproteome.5b00692 J. Proteome Res. 2015, 14, 5263−5272

Article

Journal of Proteome Research suppressor protein is required for proper assembly of an extracellular fibronectin matrix. Mol. Cell 1998, 1, 959−968. (29) Clifford, S. C.; Cockman, M. E.; Smallwood, A. C.; Mole, D. R.; Woodward, E. R.; Maxwell, P. H.; Ratcliffe, P. J.; Maher, E. R. Contrasting effects on HIF-1alpha regulation by disease-causing pVHL mutations correlate with patterns of tumourigenesis in von HippelLindau disease. Hum. Mol. Genet. 2001, 10, 1029−1038. (30) Hoffman, M. A.; Ohh, M.; Yang, H.; Klco, J. M.; Ivan, M.; Kaelin, W. G., Jr. von Hippel-Lindau protein mutants linked to type 2C VHL disease preserve the ability to downregulate HIF. Hum. Mol. Genet. 2001, 10, 1019−1027. (31) Kurban, G.; Hudon, V.; Duplan, E.; Ohh, M.; Pause, A. Characterization of a von Hippel Lindau pathway involved in extracellular matrix remodeling, cell invasion, and angiogenesis. Cancer Res. 2006, 66, 1313−1319. (32) Hosoya, N.; Sakumoto, M.; Nakamura, Y.; Narisawa, T.; Bilim, V.; Motoyama, T.; Tomita, Y.; Kondo, T. Proteomics identified nuclear N-myc downstream-regulated gene 1 as a prognostic tissue biomarker candidate in renal cell carcinoma. Biochim. Biophys. Acta, Proteins Proteomics 2013, 1834, 2630−2639. (33) Yamasaki, T.; Seki, N.; Yoshino, H.; Itesako, T.; Yamada, Y.; Tatarano, S.; Hidaka, H.; Yonezawa, T.; Nakagawa, M.; Enokida, H. Tumor-suppressive microRNA-1291 directly regulates glucose transporter 1 in renal cell carcinoma. Cancer science 2013, 104, 1411. (34) Suganuma, N.; Segade, F.; Matsuzu, K.; Bowden, D. W. Differential expression of facilitative glucose transporters in normal and tumour kidney tissues. BJU Int. 2007, 99, 1143−1149. (35) Chan, D. A.; Sutphin, P. D.; Nguyen, P.; Turcotte, S.; Lai, E. W.; Banh, A.; Reynolds, G. E.; Chi, J. T.; Wu, J.; Solow-Cordero, D. E.; Bonnet, M.; Flanagan, J. U.; Bouley, D. M.; Graves, E. E.; Denny, W. A.; Hay, M. P.; Giaccia, A. J. Targeting GLUT1 and the Warburg effect in renal cell carcinoma by chemical synthetic lethality. Sci. Transl. Med. 2011, 3, 94ra70. (36) Behrooz, A.; Ismail-Beigi, F. Dual control of glut1 glucose transporter gene expression by hypoxia and by inhibition of oxidative phosphorylation. J. Biol. Chem. 1997, 272, 5555−5562. (37) Airley, R.; Loncaster, J.; Davidson, S.; Bromley, M.; Roberts, S.; Patterson, A.; Hunter, R.; Stratford, I.; West, C. Glucose transporter glut-1 expression correlates with tumor hypoxia and predicts metastasis-free survival in advanced carcinoma of the cervix. Clin. Cancer Res. 2001, 7, 928−934. (38) Kim, D. S.; Choi, Y. P.; Kang, S.; Gao, M. Q.; Kim, B.; Park, H. R.; Choi, Y. D.; Lim, J. B.; Na, H. J.; Kim, H. K.; Nam, Y. P.; Moon, M. H.; Yun, H. R.; Lee, D. H.; Park, W. M.; Cho, N. H. Panel of candidate biomarkers for renal cell carcinoma. J. Proteome Res. 2010, 9, 3710− 3719. (39) Grupp, K.; Sanader, S.; Sirma, H.; Simon, R.; Koop, C.; Prien, K.; Hube-Magg, C.; Salomon, G.; Graefen, M.; Heinzer, H.; Minner, S.; Izbicki, J. R.; Sauter, G.; Schlomm, T.; Tsourlakis, M. C. High lysophosphatidylcholine acyltransferase 1 expression independently predicts high risk for biochemical recurrence in prostate cancers. Mol. Oncol. 2013, 7, 1001−1011. (40) Mardilovich, K.; Shaw, L. M. Hypoxia regulates insulin receptor substrate-2 expression to promote breast carcinoma cell survival and invasion. Cancer Res. 2009, 69, 8894−8901. (41) Pankratz, S. L.; Tan, E. Y.; Fine, Y.; Mercurio, A. M.; Shaw, L. M. Insulin receptor substrate-2 regulates aerobic glycolysis in mouse mammary tumor cells via glucose transporter 1. J. Biol. Chem. 2009, 284, 2031−2037. (42) Kim, J. W.; Tchernyshyov, I.; Semenza, G. L.; Dang, C. V. HIF1-mediated expression of pyruvate dehydrogenase kinase: a metabolic switch required for cellular adaptation to hypoxia. Cell Metab. 2006, 3, 177−185. (43) Papandreou, I.; Cairns, R. A.; Fontana, L.; Lim, A. L.; Denko, N. C. HIF-1 mediates adaptation to hypoxia by actively downregulating mitochondrial oxygen consumption. Cell Metab. 2006, 3, 187−197. (44) Vanharanta, S.; Shu, W.; Brenet, F.; Hakimi, A. A.; Heguy, A.; Viale, A.; Reuter, V. E.; Hsieh, J. J.; Scandura, J. M.; Massague, J.

Epigenetic expansion of VHL-HIF signal output drives multiorgan metastasis in renal cancer. Nat. Med. 2013, 19, 50−56. (45) Maxwell, P. H.; Wiesener, M. S.; Chang, G. W.; Clifford, S. C.; Vaux, E. C.; Cockman, M. E.; Wykoff, C. C.; Pugh, C. W.; Maher, E. R.; Ratcliffe, P. J. The tumour suppressor protein VHL targets hypoxiainducible factors for oxygen-dependent proteolysis. Nature 1999, 399, 271−275. (46) Zimmer, M.; Ebert, B. L.; Neil, C.; Brenner, K.; Papaioannou, I.; Melas, A.; Tolliday, N.; Lamb, J.; Pantopoulos, K.; Golub, T.; Iliopoulos, O. Small-molecule inhibitors of HIF-2a translation link its 5′UTR iron-responsive element to oxygen sensing. Mol. Cell 2008, 32, 838−848. (47) An, J.; Fisher, M.; Rettig, M. B. VHL expression in renal cell carcinoma sensitizes to bortezomib (PS-341) through an NF-kappaBdependent mechanism. Oncogene 2005, 24, 1563−1570. (48) An, J.; Rettig, M. B. Mechanism of von Hippel-Lindau proteinmediated suppression of nuclear factor kappa B activity. Mol. Cell. Biol. 2005, 25, 7546−7556. (49) Cummins, E. P.; Berra, E.; Comerford, K. M.; Ginouves, A.; Fitzgerald, K. T.; Seeballuck, F.; Godson, C.; Nielsen, J. E.; Moynagh, P.; Pouyssegur, J.; Taylor, C. T. Prolyl hydroxylase-1 negatively regulates IkappaB kinase-beta, giving insight into hypoxia-induced NFkappaB activity. Proc. Natl. Acad. Sci. U. S. A. 2006, 103, 18154− 18159. (50) Sarada, S.; Himadri, P.; Mishra, C.; Geetali, P.; Ram, M. S.; Ilavazhagan, G. Role of oxidative stress and NFkB in hypoxia-induced pulmonary edema. Exp. Biol. Med. (London, U. K.) 2008, 233, 1088− 1098. (51) Sourbier, C.; Danilin, S.; Lindner, V.; Steger, J.; Rothhut, S.; Meyer, N.; Jacqmin, D.; Helwig, J. J.; Lang, H.; Massfelder, T. Targeting the nuclear factor-kappaB rescue pathway has promising future in human renal cell carcinoma therapy. Cancer Res. 2007, 67, 11668−11676. (52) Qi, H.; Ohh, M. The von Hippel-Lindau tumor suppressor protein sensitizes renal cell carcinoma cells to tumor necrosis factorinduced cytotoxicity by suppressing the nuclear factor-kappaBdependent antiapoptotic pathway. Cancer Res. 2003, 63, 7076−7080. (53) Peruzzi, B.; Athauda, G.; Bottaro, D. P. The von Hippel-Lindau tumor suppressor gene product represses oncogenic beta-catenin signaling in renal carcinoma cells. Proc. Natl. Acad. Sci. U. S. A. 2006, 103, 14531−14536. (54) Wisniewski, J. R.; Zougman, A.; Nagaraj, N.; Mann, M. Universal sample preparation method for proteome analysis. Nat. Methods 2009, 6, 359−362. (55) Omerovic, J.; Hammond, D. E.; Prior, I. A.; Clague, M. J. Global snapshot of the influence of endocytosis upon EGF receptor signaling output. J. Proteome Res. 2012, 11, 5157−5166. (56) Olsen, J. V.; Blagoev, B.; Gnad, F.; Macek, B.; Kumar, C.; Mortensen, P.; Mann, M. Global, in vivo, and site-specific phosphorylation dynamics in signaling networks. Cell 2006, 127, 635−648. (57) Cox, J.; Mann, M. MaxQuant enables high peptide identification rates, individualized p.p.b.-range mass accuracies and proteome-wide protein quantification. Nat. Biotechnol. 2008, 26, 1367−1372.

5272

DOI: 10.1021/acs.jproteome.5b00692 J. Proteome Res. 2015, 14, 5263−5272